DEPARTMENT OF EXPERIMENTAL EVOLUTION. 139 



There are several reasons for not realizing a very close relation 

 between observed and expected; but the chief one is that there are 

 sometimes several ''expected" results; and in these cases the most 

 varied result was taken as the "expected." Nevertheless, the relation 

 between the two series is fairly close, the greatest difference being an 

 unexpected excess of choleric-melancholies and corresponding defi- 

 ciency of choleric-phlegmatics ; also an excess of calm-cheerful and 

 deficiency of calm-phlegmatic. These discrepancies imply a rather 

 shght error in the classification of the observed cases. 



We conclude, then, from our own data that the hypothesis is con- 

 firmed. In addition, an examination of the hterature reveals clear 

 evidence that a difference in the inheritance of extreme hyperkinesis 

 (a dominant) and extreme hypokinesis (a recessive) has unconsciously 

 been observed. And the differences in the conclusions of Rosanoff 

 and Orr^ and of Riidin^ concerning the inheritance of manic-depressive 

 insanity — one regarding it as recessive and the other regarding it as 

 sometimes dominant — are easily explained on the ground of its com- 

 plex hereditary nature. 



Finally the study throws hght upon the "springs of conduct." Just 

 what we shall do. in any situation, is determined by numerous factors, 

 but the general nature of our reactions, whether violent or repressed^ 

 is determined by the hereditary nature of our temperaments. The 

 romantic and the classic type of reacting, the hyperkinetic and the 

 hypokinetic, the radical and the conservative, the feebly inhibited and 

 the strongly inhibited, constitute a duahsm that runs through our whole 

 population. 



(3) Nomadism. — A study was made of the records of 100 families 

 one or more members of which have shown a tendency to run away 

 from home, families, and duties or to engage in a nomadic occupation. 

 In these families, out of 186 nomadics only 15 are females. The hy- 

 pothesis that best fits the facts is that nomadism is a sex-linked trait. 

 There is no clear case of a nomadic daughter whose father is known to 

 be non-nomadic; all daughters (4) of two nomadic parents are nomadic. 

 The criterion that half of the daughters and half of the sons of nomad- 

 bearing fraternities derived from nomadic fathers are nomadic is satis- 

 fied so far as sons go, but the number of nomadic daughters is less than 

 expected (only 26 instead of 50 per cent are nomadic). In general,, 

 our histories show that nomadic fathers may have no nomadic sons, 

 but there is no case of a nomadic mother of more than two children 

 none of whom is nomadic. Thus the hypothesis is supported with the 

 proviso that certain cultural conditions may have a repressive effect 

 even in the absence of natural inhibitions. 



^Rosanoff, A. J., and Florence Orr, A study of insanity in the light of the Mendelian theory. 

 Amer. J. Insanity, 68, 221-261 (1911) ; also Eugenics Records Office Bull., No. 5. 



^Riidin, E., Einige Wege und Ziele der Familienforschung, mit Riicksicht auf die Psychiatrie,. 

 Zs. ges. Neurol. Psychiat., 7 (Hft. 5), 487-585 (1911). 



